>) 
34 MEMOIRS OF THE CARNEGIE MUSEUM 
verse processes instead of being horizontal are directed successively more and more 
obliquely upward as in the succeeding dorsals. 
The Fifth, Sixth, Seventh, Kighth, Ninth, Tenth and Eleventh Dorsals (Plate I1., Figs. 
5-11, Series 1 and 2).— The neural arches and spines of all these vertebrae were 
found interlocked by their zygapophyses as shown in the diagram from 14-20 in- 
clusive. The centra of the fifth and sixth had become detached and lay as shown 
at 14’ and 15’. The remaining centra were in position at the base of their respec- 
tive spines. The neural spines, transverse processes, capitular rib facets, neural 
arches, etc., form a regularly graudated series except that the capitular rib facet of the 
sixth is much larger than in the other vertebree. The neural arch, spine and trans- 
verse processes of the fifth were much injured, but the spine is nearly entire and it 
is evident that it pertained to the vertebra immediately posterior to that just 
described as indicated also by the centrum. The spine is still compressed antero- 
posteriorly but decidedly deeper in that direction than the spine of the verte- 
bra just described. In the spine of the succeeding or sixth dorsal the transverse 
and anteroposterior diameters are subequal. A hyposphene-hypantrum articula- 
tion begins with the sixth dorsal and continues throughout the remaining dorsal 
series. 
The Twelfth, Thirteenth, and Fourteenth Dorsals (Plate I1., Figs. 12-14, Series 1). — 
These vertebree do not differ materially from the same vertebree already described as 
pertaining to the type of H. priscus. They were found as shown at 21, 22 and 23 
in the diagram, interposed between the series just described and the anterior ex- 
tremity of the sacrum. As shown in the diagram the neural arches were in posi- 
tion relative to one another but the centra were a little removed from their normal 
positions. They are all in a nearly perfect condition. 
If the reader has followed carefully the above description of the dorsals pertain- 
ing to the present skeleton together with those which pertained to the type of H. pris- 
cus and will examine the accompanying figures it will have become apparent that 
the complete dorsal series in Haplocanthosawrus must have consisted of not less than 
fourteen free vertebree while it is scarcely possible that there were more than four- 
teen. This is a very marked increase over the number (ten) which is believed to 
have formed the complement of free dorsals in Diplodocus, Brontosaurus and Moro- 
suurus. Nor does this increase in the number of dorsals in the present genus seem 
to have been made at the expense of the cervical series, for as near as we can judge, 
Haplocanthosauwrus, like Diplodocus, was provided with fifteen cervicals. Our deter- 
mination of the number of cervicals however does not rest on anything like so good 
a basis as does our determination of the number of dorsals but there can be little 
